BEGIN:VCALENDAR
VERSION:2.0
PRODID:-//Talks.cam//talks.cam.ac.uk//
X-WR-CALNAME:Talks.cam
BEGIN:VEVENT
SUMMARY:Subsurface imaging with deep learning - Song Hou -- CGG
DTSTART:20191010T120000Z
DTEND:20191010T133000Z
UID:TALK122974@talks.cam.ac.uk
CONTACT:James Fergusson
DESCRIPTION:To see what is underneath the surface of the earth\, we can dr
 ill a borehole\, or perform what is called “seismic imaging”. Whereas 
 boreholes give very localized information\, seismic images can result in t
 ens of thousands square kilometres being digitally imaged down to several 
 kilometres\, and analysed for rock strata\, structures and fluids such as 
 oil and gas.  Such experiments use multi-sensor arrays to listen to the en
 ergy reflected from the earth’s interior much like a bat uses ultrasonic
  echo-location or a CT scanner uses X-rays to image inside our bodies. The
  physics of wave-propagation is well understood\, in theory\, and yet seis
 mic imaging is a hard problem\, computationally intensive\, and often the 
 solution is under-determined\, despite TBytes of field data available for 
 analysis. Could deep learning be a game changer?
LOCATION:Kavli Large Meeting Room\, Kavli Building
END:VEVENT
END:VCALENDAR